Seventy per cent of digital transformations fail. McKinsey puts the figure between 70% and 90%. Bain's 2024 analysis found that 88% of business transformations fail to achieve their original ambitions [1][2]. These are not fringe studies — they are the consensus from the world's most respected consulting firms, repeated across two decades of research.
And yet companies keep launching digital transformation initiatives the same way: pick a technology, buy licences, hire consultants, and hope that the new system imposes order on chaos. It does not. The technology amplifies whatever already exists. If your processes are broken, your shiny new platform will break them faster, at greater expense, and with less visibility into what went wrong.
The 70% failure rate is not a technology problem. It is a sequencing problem. Companies are doing the right things in the wrong order — and there are three patterns that account for the vast majority of failures.
What Are the Three Failure Patterns Killing Digital Transformations?
Every failed digital transformation we have examined falls into one of three patterns. Each starts with good intentions and ends with wasted budget, frustrated teams, and a quiet return to spreadsheets.
Pattern 1: The ERP Graveyard. A growing business reaches a point where its cobbled-together systems can no longer keep up. Someone — usually a board member or a well-meaning advisor — suggests an ERP. The company commits six or seven figures, spends 12 to 24 months on implementation, and discovers that the ERP requires the business to conform to its logic rather than the other way around. The statistics are damning: 55-75% of ERP implementations fail to meet their objectives, with average cost overruns of 189% [3]. Ninety-five per cent of companies that fail dedicate less than 10% of total budget to education, training, and change management [3]. The ERP is not the villain — the sequencing is. These businesses automated processes they had never properly mapped, simplified, or validated.
Pattern 2: Zapier Spaghetti. At the other end of the spectrum sits the scrappy, tech-savvy founder who discovers no-code automation tools and connects everything to everything. Zapier to sync the CRM with the project management tool. Make (formerly Integromat) to push invoice data into the accounting system. A Google Sheet acting as a pseudo-database. It works — until it does not. When one step in a multi-step workflow fails, the entire automation stops, leaving subsequent steps unprocessed and creating data inconsistencies [4]. The automation map looks like a plate of spaghetti, and changing one workflow means updating five others. The business has replaced manual chaos with automated chaos — and nobody understands the full picture any more.
Pattern 3: The Shadow IT Crisis. When the official systems do not work, people build their own. Employees use SaaS applications without IT approval in 80% of cases [5]. The average enterprise believes it uses 91 cloud services; actual usage averages 1,220 [5]. Teams create private Notion workspaces, personal Trello boards, WhatsApp groups that hold critical project decisions, and Google Sheets that serve as the actual source of truth for processes the official system was supposed to manage. By 2027, Gartner predicts 75% of employees will use technology outside of IT oversight [5]. Shadow IT is not disobedience — it is a rational response to systems that do not serve the people using them.
All three patterns share the same root cause: the business tried to solve a process problem with a technology decision.
Why Does Technology-First Thinking Guarantee Failure?
Because technology is an amplifier, not a fixer. If you automate a broken process, you get a broken process that runs at machine speed. If you integrate systems around a workflow that nobody has simplified, you get complex integrations that are expensive to maintain and impossible to debug. If you deploy an enterprise platform without first understanding how work actually flows through your organisation, you get a platform that people work around rather than within.
McKinsey's research is unambiguous on this point: organisations that invest in cultural change and process redesign see 5.3 times higher success rates than those focused only on technology [1]. The main culprits behind failure include lack of clear goals, weak change management, inadequate tracking of progress, and failure to prioritise people as part of the transformation journey [1].
The uncomfortable truth is that most digital transformation budgets are spent on the wrong layer. The technology is the easiest part. The hard part — and the part that determines success or failure — is the operational architecture underneath: how work actually flows, who owns each handoff, where decisions get made, and what happens when exceptions arise. Without that foundation, any technology you deploy is built on sand.
Is Your Business Operations Tech Stack Actually Making Things Worse?
Here is a diagnostic question that reveals more than most audits: how many tools does your team use to complete a single end-to-end process?
If the answer is more than three, and information has to be manually transferred between them, your tech stack is not a solution — it is part of the problem. Every manual handoff between systems is a point where errors enter, delays accumulate, and someone's time gets consumed by work that adds no value.
The pattern is particularly acute in small and mid-market businesses. The founder or operations lead, seeking process management software for the business, evaluates tools in isolation. A CRM here. A project management tool there. An invoicing platform. A communication tool. Each solves a narrow problem. None of them talk to each other in a meaningful way. The result: too many business tools, not integrated, and a team that spends more time managing tools than doing actual work.
The no-code tools for business operations — Zapier, Make, Airtable, Notion — are genuinely powerful. But they are components, not architecture. Using them without first designing the underlying process is like buying excellent building materials without an architectural plan. You end up with a structure that sort of works, cannot be easily modified, and collapses under stress.
What Should Businesses Do Instead of Leading with Technology?
The answer is deceptively simple: fix the sequence. The ESIA framework — Eliminate, Simplify, Integrate, Automate — provides the correct order of operations, and it traces its lineage to the Toyota Production System [6].
Eliminate first. Before improving any process, ask: what would happen if we stopped doing this entirely? Many process steps exist because they were created years ago for a problem that no longer exists — an approval layer added during a crisis and never removed, a report nobody reads, a data-collection step that feeds a dashboard no one checks. Every step you eliminate is a step that never needs documentation, training, tooling, or maintenance.
Simplify what remains. Fewer handoffs. Fewer approval layers. Fewer steps between input and output. If a process requires a 20-step checklist, the process is too complex — and complexity is where errors breed. Simplification is not about cutting corners. It is about removing unnecessary friction so that essential work flows cleanly.
Integrate the data flows. Connect your systems so that information flows without manual re-entry. This is where process management software and no-code tools genuinely shine — but only when applied to a process that has already been cleaned up. Integration built on top of a simplified process is elegant. Integration built on top of a messy one is expensive spaghetti.
Automate last. Only what is stable, rule-based, and high-volume. Automation is the final layer, not the first. It locks in the gains from the previous three steps. Applied to a clean foundation, automation is transformative. Applied to a messy one, it is the 70% failure statistic in action.
The order is not optional. Each step depends on the one before it. You cannot meaningfully simplify a process until you have eliminated the unnecessary parts. You cannot integrate around a process that is still overly complex. And you should not automate what has not been stabilised.
How Does This Work in Practice?
Consider a real example. One client came to Alcara Partners with a core operational process that required 20 people working for 6 months to complete. It consumed enormous resources and still produced a backlog worth over 900,000 euros.
We did not start by recommending software. We started by mapping the process as it actually existed — not the org chart version, but how work truly moved through the business day to day. Then we applied ESIA:
We eliminated roughly 40% of process steps — approvals that added no value, data collection that fed reports nobody used, handoffs that existed because of an old organisational structure that had since changed. We simplified the remaining steps by reducing handoff points, consolidating decision authority, and streamlining the workflow from a branching maze into a linear sequence with clear exception paths. We integrated the relevant systems so that data entered once propagated automatically. And only then did we build the automation layer — targeting the high-volume, rule-based tasks that now sat on a clean, stable foundation.
The result: 1 person, 1 week. The 900,000-euro backlog was cleared in 3 months. Not because of a brilliant piece of software, but because the process underneath was redesigned before any technology was applied.
That is the difference between a digital transformation that works and one that joins the 70%.
At Alcara Partners, we start every engagement with the Alcara Diagnostic: a focused operational assessment that maps how your business actually runs, identifies friction points, and quantifies what they cost you in euros and hours. We do not sell software. We redesign the operational architecture — and only then help you select and implement the right tools for the redesigned process.
If your business has broken processes, too many tools that are not integrated, or a digital transformation that has stalled, the problem is almost certainly not the technology. It is the sequence. And the sequence is fixable.
Frequently Asked Questions
Why do 70% of digital transformations fail?
The primary cause is sequencing: businesses invest in technology before redesigning the processes that technology is meant to support. McKinsey and BCG research consistently shows that failures stem from people-centric issues — unclear vision, poor change management, lack of user adoption, and cultural resistance — rather than the technology itself [1][2]. When you automate a broken process, you get an expensive broken process.
Is Zapier not enough for business processes?
Zapier and similar no-code tools are powerful for point-to-point integrations, but they are components, not architecture. When workflows grow complex, they create fragile chains where a single failure stops everything downstream and creates data inconsistencies [4]. The real issue is not the tool — it is using automation tools before the underlying process has been mapped, simplified, and stabilised.
What is shadow IT and why does it matter for digital transformation?
Shadow IT refers to technology that employees adopt without official approval — personal Trello boards, WhatsApp groups for project decisions, private spreadsheets as the actual source of truth. Eighty per cent of SaaS usage happens without IT approval [5]. Shadow IT is a symptom, not a cause: when official systems do not serve the people using them, those people find alternatives. Fixing the process reduces the incentive for workarounds.
What is the ESIA framework?
ESIA stands for Eliminate, Simplify, Integrate, Automate — in that exact order. It is a process redesign methodology rooted in the Toyota Production System [6]. The framework ensures you remove waste and reduce complexity before building integrations or automation. The sequence is critical: each step reduces the scope and complexity for the next.
Why do ERP implementations fail so often?
Between 55% and 75% of ERP implementations fail to meet their objectives, with cost overruns averaging 189% [3]. The core issue is that businesses treat ERP as a solution rather than a platform. Without first mapping and redesigning their processes, companies force-fit their operations into the ERP's logic — and then wonder why adoption is low and workarounds proliferate.
How do I know if my business has a process problem or a technology problem?
Ask three questions. Are your teams re-entering data between systems? Do new hires take months to become productive? Does output quality vary depending on who does the work? If the answer to any of these is yes, you have a process problem — and adding more technology will make it worse, not better.
What should I do before starting a digital transformation?
Start with an operational diagnostic. Map how your business actually runs — not the idealised version, but the reality. Identify where handoffs break, where information gets stuck, and where people are doing work that should not exist. Then apply the ESIA sequence: eliminate unnecessary steps, simplify what remains, integrate data flows, and automate last. This foundation determines whether your transformation succeeds or joins the 70%.
Can small businesses benefit from the ESIA framework?
Absolutely. ESIA is not enterprise methodology — it is a sequencing principle. Small businesses with 10 to 50 employees often benefit the most because they can move faster, have fewer legacy systems to untangle, and see the impact of process redesign almost immediately. The framework works at any scale because the underlying logic is universal: remove waste before you optimise.
References
[1] McKinsey & Company. "Why do most transformations fail? A conversation with Harry Robinson." https://www.mckinsey.com/capabilities/transformation/our-insights/why-do-most-transformations-fail-a-conversation-with-harry-robinson
[2] Bain & Company / BCG. "Flipping the Odds of Digital Transformation Success." https://www.bcg.com/publications/2020/increasing-odds-of-success-in-digital-transformation
[3] Panorama Consulting Group / Godlan. "ERP Implementation Failure Statistics: 2025 Research." https://godlan.com/erp-implementation-failure-statistics/
[4] Autonoly. "Why Your Zapier Workflows Keep Breaking: 7 Fixes for Unreliable Automation." https://www.autonoly.com/blog/689c0d2be633225ff19e1004/why-your-zapier-workflows-keep-breaking-7-fixes-for-unreliable-automation
[5] ElectroIQ / Zluri. "Shadow IT Statistics You Need to Know Now (2026)." https://electroiq.com/stats/shadow-it-statistics/
[6] Toyota Europe. "Toyota Production System." https://www.toyota-europe.com/about-us/toyota-vision-and-philosophy/toyota-production-system
[7] Alcara Partners. Internal data from operational diagnostics 2024-2026. Businesses with five to fifteen million euros in revenue.